Raster maps with mapsf

mapsf focuses on creating thematic maps based on vector data (maps are generated from sf objects). However, it is also possible to map raster data thanks to the mf_raster() function.

The latest update of mapsf (v0.9.0) expands the mapping capabilities for raster data. Emulating the functionality of terra (and leveraging its plotting function internally), mf_raster() now provides three types of maps: “classes,” “continuous,” and “interval.”

Classes

This type of raster map is suitable for rasters containing qualitative data (categories).
In this example we display a map of Corine Land Cover classes over the Lot region in France.

library(mapsf)
# import raster
clc <- terra::rast("clc_46.tif")
# import vector (Lot region polygon)
dep <- sf::st_read("dep.gpkg", "dep", quiet = TRUE) 

# define a color palette
clc_pal <- c("#E6004D", "#FF0000", "#CC4DF2", "#CC0000", "#E6CCE6", "#A600CC", 
             "#FFE6FF", "#FFFFA8", "#E68000", "#F2A64D", "#E6E64D", "#FFE64D", 
             "#E6CC4D", "#80FF00", "#00A600", "#4DFF00", "#CCF24D", "#A6FF80", 
             "#A6E64D", "#A6F200", "#00CCF2", "#80F2E6")

# map the raster
mf_raster(clc, pal = clc_pal, expandBB = c(0, 0.5, 0.01, 0),
          leg_title = "", leg_size = .7, leg_box_cex = c(.6, .85), leg_pos = "topleft")

# add region borders 
mf_map(dep, col = NA, lwd = 2, add = TRUE)

# add map layout elements
mf_title("CORINE Land Cover - Lot region - 2018", tab = FALSE, pos = "center") 
mf_credits("T. Giraud, 2024 | EEA 2018")
mf_arrow(pos = "topright")
mf_scale(10)

Continuous

This type of map is appropriate for numeric rasters.
In the following example we display elevation maps of the Lot region.

# import raster
elev <- terra::rast("elev_46.tif")

# map the raster
mf_raster(elev, leg_pos = "topleft")

# add layout elements
mf_title("Elevation - Lot region") 
mf_credits("SRTM 90m DEM Digital Elevation Database, 2018")
mf_arrow(pos = "topright")
mf_scale(5)

The colors are set using a palette name or a vector of colors to map the continuous values.
See ?hcl.colors() examples to check palette names and colors.

# select a palette with a palette name
mf_raster(elev, pal = "Rocket", expandBB = c(0,0,0,.3))
# define a palette with colors (from red to black)
mf_raster(elev, pal = c("#fe5234", "#000000"), expandBB = c(0,0,0,.3))

It is also possible to give break values and corresponding colors. Breaks are inflection points of the color scale.

library(terra)

terra 1.7.65

# set a cartographic theme
mf_theme("green", bg = "black")

# map the raster
mf_raster(elev, breaks = c(67, 150, 550, 776), 
          pal = c("white", "khaki", "darkgreen", "black"),
          leg_val_rnd = 0, leg_pos = "topleft", leg_box_cex = c(.65, 1),
          leg_title = "Elevation (m)", leg_horiz = TRUE, leg_frame = TRUE, 
          leg_frame_border = NA, leg_bg = "grey20")

# add an inset 
mf_inset_on(x = "worldmap", pos = "bottomright", cex = .2)
mf_worldmap(x = elev)
mf_inset_off()

# add layout elements
mf_title("Elevation - Lot region", tab = FALSE, pos = "center") 
mf_credits("T. Giraud, 2024 | SRTM 90m DEM Digital Elevation Database, 2018", 
           bg = "#000000CC")
mf_arrow(pos = "topright")
mf_scale(5)

# add annotations
max_elev_coord <- xyFromCell(elev, where.max(elev)[2])[1,]
mf_annotation(x = max_elev_coord, "Highest Point", 
              col_arrow = "white", col_txt = "white", 
              pos = "bottomleft", halo = TRUE, cex = 1.2)

Interval

This type of map is also appropriate for numeric rasters. The numeric values are classified using one the methods provided by mf_get_breaks().

# set a cartographic theme
mf_theme("default")

# map the raster
mf_raster(elev, type = "interval", 
          breaks = "q6", pal = "Turku", rev = TRUE,
          leg_val_rnd = 1, leg_pos = "topleft", leg_title = "Elevation (m)")

# add layout elements
mf_title("Elevation - Lot region", tab = FALSE, pos = "center") 
mf_credits("SRTM 90m DEM Digital Elevation Database, 2018", 
           pos = "rightbottom")
mf_arrow(pos = "topright")
mf_scale(5, pos = "bottomleft")

Multi band

The last kind of map that mf_raster() can take care of is color composite maps using three bands.

library(maptiles)
dep <- sf::st_transform(dep, "EPSG:3857")
# get an OpenStreetMap image on the region
(osm <- get_tiles(x = dep, provider = "OpenStreetMap", crop = TRUE, zoom = 9))

class       : SpatRaster 
dimensions  : 433, 448, 3  (nrow, ncol, nlyr)
resolution  : 305.7481, 305.7481  (x, y)
extent      : 109152.1, 246127.2, 5496740, 5629129  (xmin, xmax, ymin, ymax)
coord. ref. : WGS 84 / Pseudo-Mercator (EPSG:3857) 
source(s)   : memory
colors RGB  : 1, 2, 3 
names       : lyr.1, lyr.2, lyr.3 
min values  :     3,    54,     3 
max values  :   251,   249,   244 

# set them margins
mf_theme(mar = c(0,0,0,0))

# display the map
mf_raster(osm)
mf_map(dep, add = TRUE, col = NA, lwd = 3)

# add attribution
mf_credits(get_credit("OpenStreetMap"))